GB2218252A

GB2218252A - Sound synthesizer

Info

Publication number: GB2218252A
Application number: GB8904224A
Authority: GB
Inventors: Masami Katsui; Yoshiyuki Terashima
Original assignee: Seiko Epson Corp
Current assignee: Seiko Epson Corp
Priority date: 1988-03-03
Filing date: 1989-02-24
Publication date: 1989-11-08
Anticipated expiration: 2009-02-24
Also published as: JP2661211B2; GB9203246D0; GB2251971A; GB2251513A; GB9203224D0; HK171695A; GB8904224D0; JPH01315792A; GB2218252B; GB2251513B; GB2251971B

Abstract

A sound synthesizer comprises a main store for data providing melody information, and at least one sound source (3) controlled by data signals output from the main store for generating an output representing melodic sound. The sound source comprises a further store (61) for data regarding the timbre of a sound, and a pitch control circuit (59, 60) which, in response to the respective data signal representing sound pitch output from the mains store, is arranged to co-operate with the further store so as to generate a sound quality signal representing sound of a selected pitch and timbre. The sound source further comprises a sound control circuit (55, 56, 57, 58, 63, 65) responsive to the sound quality signal and to at least one further data signal representing a further sound characteristic (e.g. volume or envelope shape) output from the main store for providing the output from the sound source. A further sound source generates an output representing rhythmic sound and comprises a noise generating circuit and two bell sound generating circuits. <IMAGE>

Description

SOUND SYNTHESIZER The present invention relates to a sound synthesizer.

A conventional sound synthesizer is shown in 'Figure 9. In this synthesizer, a clock signal output from an oscillating circuit 91 is variably divided by a note length generating circuit 92. A main ROM 93 serves as a store for data on the length and the pitch of notes in a melody. From the data on the length of notes read from the main ROM 93, a division ratio is determined for the note length generating circuit 92. The thus divided clock signal is input to a main counter 95, which increments the read address in the main ROM 93 in accordance with the length of each note. Likewise, data on the note pitch read from the main ROM 93 deter mines a division ratio for a note pitch generating circuit 94.The note pitch generating circuit 94 variably divides the clock signal supplied from the oscillating circuit 91 in accordance with this deter mined division ratio so that a clock signal having a frequency in accordance with the note pitch is output.

An envelope is added to this clock signal in an envelope generating circuit 96. The envelope generating circuit 96 comprises a capacitor and a resistor so that a pre determined analog waveform is formed by discharging the charge stored in the capacitor via the resistor on each occasion. The signal from the envelope generating circuit 96 is supplied to a speaker and a tone having a note pitch determined by data stored in the main ROM 93 is generated for a time period corresponding to the length of the note. By successively reading data from the main ROM 93, a melody can be automatically generated.

This conventional sound synthesizer deals solely with sound represented by rectangular waveforms and with envelopes in the form of an exponential curve as formed by the capacitor - resistor circuit of the envelope generating circuit 96. Therefore, a uniform tone of poor quality is produced such that the conventional synthesizer has been used only for call holding on a telephone, melody cards or the like. No sound accentuation is present, the number of sound sources is limited, the generation of rhythmic sound is impossible, and natural, stereo, and profound sound cannot be generated.

It is an object of the present invention, at least in its preferred form, to provide a sound synthesizer capable of generating sound of a variety of qualities, including rhythmic sounds.

Although the present invention is primarily directed to any novel integer or step, or combination of integers or steps, herein disclosed and/or as shown in the accompanying drawings, nevertheless, according to one particular aspect of the present invention to which, however, the invention is in no way restricted, there is provided a sound synthesizer comprising a main store for data providing melody information, and a sound source controlled by data signals output from the main store for generating an output representing melodic sound, the sound source comprising a further store for data regarding the timbre of a sound, a pitch control circuit which, in response to a respective data signal representing sound pitch output from the main store, is arranged to co-operate with the further store so as to generate a sound quality signal representing sound of a selected pitch and timbre, and a sound control circuit responsive to the sound quality signal and to at least one further data signal representing a further sound characteristic output from the main store for providing the output.

In the preferred embodiment, the further store contains data on the waveform of the sound to be produced, and such data is repeatedly read out from the further store at a frequency determined by the data signal representing sound pitch. In addition, the sound source contains another store containing data on an envelope for the waveform, and such data is supplied to a digital to analog converter for conversion to an analog voltage. The analog voltage is then combined with a digital signal representing the waveform to produce an output for supply to a speaker. Thus, sound having an optional timbre or tone in accordance with the selected waveform and an optional lingering quality in accordance with the envelope can be obtained.

Preferably, the synthesizer includes a plurality of the sound sources, and it may also include a source of rhythmic sound.

According to another aspect of the invention, there is provided a synthesizer comprising a main store for data providing melody information, and a source of rhythmic sound controlled by data signals output from the main store for generating an output representing rhythmic sound, the rhythmic sound source including means for generating a rhythm signal in the form of pulses, and a rhythmic sound control circuit arranged in response to the rhythm signal and the data signals output from the main store to generate the output representing rhythmic sound.

A further aspect of the invention also features a sound synthesizer comprising a main store for data providing melody information, and a sound source controlled by data signals output from the main store for generating an output representing melodic sound, the sound source comprising means responsive to a respective one of the data signals representing sound pitch for generating a digital signal representing sound of a selected pitch, means for generating an envelope signal which is controlled in dependence upon a respective one or respective ones of the data signals representing at least one of sound tempo and sound volume, and a circuit for combining the digital signal and the envelope signal for providing the output.

The invention is described further, by way of example, with reference to the accompanying drawings, in which: Figure 1 is a diagrammatic view of a sound synthesizer embodying the present invention; Figure 2 is a block diagram of a control circuit of the sound synthesizer; Figure 3 is a block diagram of a sound source of the sound synthesizer; Figure 4 is a block diagram of a rhythm generating circuit of the sound synthesizer; Figure 5 is a block diagram showing in greater detail a portion of the circuitry illustrated in Figures 3 and 4; Figure 6 illustrates a saw shaped envelope stored in a memory of the sound source shown in Figure 3; Figure 7 illustrates a sine wave stored in a memory of the sound source shown in Figure 3; Figure 8 illustrates the relationship between some of the signals generated in the sound synthesizer in use; and Figure 9 is a block diagram of a conventional sound synthesizer.

Figure 1 shows a sound synthesizer in accordance with an embodiment of the present invention. In Figure 1, the reference numeral 1 denotes an oscillator circuit, the reference numeral 2 denotes a control circuit, the reference numeral 3 denotes a first sound source, the reference numeral 4 denotes a second sound source, the reference numeral 5 denotes a third sound source, the reference numeral 6 denotes a rhythmic sound generating circuit, and the reference numeral 7 denotes a mixing circuit.

Figure 2 illustrates an example of a circuit serving as the control circuit 2. Referring to this Figure, the reference numeral 31 denotes an input terminal to which a signal from the oscillator circuit 1 having an oscillating frequency is supplied. The reference number 32 denotes a tempo programmable counter for determining tempo by variably dividing the frequency of the signal at the terminal 31 to generate a frequency representing the timing of the shortest note in a score.

The reference numeral 33 denotes a tempo ROM, which stores tempo data for variably setting the division ratio of the tempo-programmable counter 32, and in which an address is selected by an output from a control ROM 39 for supplying the tempo data. The reference numeral 34 denotes a note programmable counter for determining the length of each note in the score by variably dividing the output from the tempo programmable counter 32. The reference numeral 35 denotes a note ROM which stores note length data for variably setting the division ratio of the note programmable counter 34 and which provides an output representing the division ratio when addressed by a main ROM 37. Thus, a clock pulse at a frequency corresponding to the desired note length is provided as an output from the note programmable counter 34.The reference numeral 36 denotes a main programmable counter for counting the clock pulses in the output from the note programmable counter 34 for each note. The main ROM 37 stores various data on the contents of the score (note length, note pitch, sound volume and sound separation of each of the sound sources, rhythm sound volume and rhythm sound separation, and jump data for executing a repetition of bars in the score), the ROM 37 being addressable by the main programmable counter 36, which is incremented by the output from the note counter 34. The reference numeral 38 denotes a control counter responsive to the jump data in the main ROM 37. The jump data serves to return the main ROM 37 to an initial address of a bar to be repeated when a repetition of a bar is required in a score.The control ROM 39 stores the address to which the main ROM 37 is to jump, the addresses in the ROM 39 being incremented by the control counter 38. The address in the main ROM 37, to which a jump is to be made, is programmed by.setting or re-setting the main programmable counter 36 in response to the data stored in the ROM 39. In addition, the control ROM 39 addresses the tempo ROM 33 so that tempo can also be changed by a jump. The reference numeral 40 denotes an output terminal for clock pulses from the tempo programmable counter 32. The reference numeral 41 denotes output terminals for score data (data on note length and pitch, sound volume and separation for the sound from each sound source, and rhythmic sound volume and separation) from the main ROM 37.The output terminals 40 and 41 are connected to respective inputs of the first, second and third sound sources 3, 4 and 5 and the rhythmic sound generating circuit 6 shown in Figure 1. The main ROM 37 stores data on the note length and pitch, -sound volume and separation of sound for the sound sources 3, 4 and 5 and data on the sound volume and separation for the rhythmic sound generating circuit 6 and it outputs such data in parallel to the terminals 41.

Figure 3 illustrates an example of a circuit serving as a respective one of the sound sources 3, 4 and 5 shown in Figure 1. Referring to this drawing, the reference numeral 51 denotes an input terminal to which an output from the tempo programmable counter 32 at the terminal 40 is supplied. The reference numeral 52 denotes an input terminal to which data on the sound volume of each note is supplied by way of one of the output terminals 41 from the main ROM 37 shown in Figure 2. The reference numeral 54 denotes an input terminal to which data on the pitch of the notes is supplied from one of the output terminals 41 shown in Figure 2. The reference numeral 66 denotes an input terminal to which data on the separation of sound is supplied from one the output terminals 41 shown in Figure 2.The reference numeral 53 denotes an input terminal for the oscillating frequency signal from the oscillator circuit 1 shown in Figure 1. The reference numeral 55 denotes an envelope ROM for storing digital data representing an envelope for a sound wave form. The reference numeral 56 denotes an envelope counter for counting the clock pulses from the tempo programmable counter 32 shown in Figure 2, that is the clock pulses whose frequency represents the timing of the shortest note in the score, for the purpose of incrementing the address of the envelope ROM 55. This counter 56 receives input signals 0 or 1 representing the separation of sound from the input terminal 66, and, when the signal is 1, indicating a separation between notes, the counter 56 is re-set and the current address of the envelope ROM 55 is made-the front most address.On the other hand, when the signal is 0; indicating the notes are connected by a tie, counting is continued and the incrementation of addresses in the envelope ROM 55 continues. The reference numeral 57 denotes a first adder circuit for adding data on the envelope for the sound waveform output from the envelope ROM 55 and data on sound volume output from the main ROM 37 shown in Figure 2, for displacing the envelope in accordance with the data on sound volume by an amount corresponding to changes in sound volume. The reference numeral 58 denotes a first DA conversion circuit for converting the digital data from the adder circuit 57 into an analog voltage, which is selected to have a value between a reference voltage and a source voltage V .The DD reference numeral 59 denotes a pitch ROM whose address is determined by data on sound pitch output from the main ROM 37. The reference numeral 60 denotes a pitch programmable counter in which the oscillating frequency of the signal from the oscillator circuit 1 is divided to produce a frequency which is N times the frequency of the pitch of the note desired and whose division ratio is determined by data stored in the pitch ROM 59. The reference numeral 61 denotes a waveform ROM for storing previously programmed data, formed by converting waveforms corresponding to desired tones (a piano sound, violin sound, or the like) into digital values. The reference numeral 62 represents a base-N counter for counting output pulses from the pitch programmable counter 60. There are a number, N, of addresses in the waveform ROM 61.Therefore, waveform data is repeatedly read out from the waveform ROM 61 at a frequency, which is the frequency of the sound pitch.

The reference numeral 63 denotes a second DA conversion circuit for converting digital data on the sound waveform output from the waveform ROM 61 into an analog voltage waveform bounded by the analog voltage values (the envelope) from the first DA conversion circuit 58 and the reference voltage. The second DA conversion circuit 63 outputs the analog voltage with the output of the first DA conversion circuit 58 as its maximum value by employing the digital. data output from the waveform ROM 61 to select the tap point in a series of potential dividers (not shown) within the DA conversion circuit 63, connected between the output of the DA conversion circuit 58 and a source of the reference voltage, from which a voltage will be taken as output from the circuit 63.The reference numeral 64 denotes an output terminal for the output from the second DA conversion circuit 63, and it serves as a final output terminal fpr an output comprising the sound wave form to which the envelope is added. The reference numeral 65 represents a switch element connected between the output terminal of the DA conversion circuit 58 and a source of the reference voltage. This switch is arranged to be switched on when data on the sound volume of the terminal 52 indicates a lack of sound, whereby the voltage input to the DA conversion circuit 63 is the reference voltage. At this time, since only the reference voltage is supplied to the DA conversion circuit 63, no operation is performed and no signal is output to the terminal 64.

Figure 4 illustrates an example of a circuit serving as the rhythmic sound generating circuit 6 shown in Figure 1. Referring to this drawing, the reference numeral 71 denotes an input terminal from the tempo programmable counter 32. The reference numeral 72 denotes an input terminal for data on the sound volume of the rhythmic sound from the main ROM 37 shown in Figure 2. The reference numeral 85 denotes an input terminal for data on the separation of the rhythmic sound from the main ROM 37 shown in Figure 2. The reference numeral 73 denotes an input terminal for the oscillating frequency signal from the oscillator circuit 1 shown in Figure 1. The reference numeral 74 denotes a rhythm envelope ROM for storing data obtained by converting an envelope for the rhythmic sound signal into a digital value.The reference numeral 75 denotes a rhythm envelope counter for incrementing the address of the rhythm envelope ROM 74 by counting the clock pulses, whose frequency represents the timing of the shortest note of the score, from the tempo programmable counter 32 shown in Figure 2. This counter 75 is arranged to receive input signals 0 or 1 representing the separation of rhythmic sound from the input terminal 85 and to be re-set when the signal is 1 so that the current address of the rhythm envelope ROM 74 is made the front most address. When the signal is 0, no resetting is performed.The reference numeral 76 denotes a second adder circuit for adding digital data on the envelope for the rhythmic sound output from the rhythm envelope ROM 74 and data on the sound volume of the rhythmic sound output from the main ROM 37 shown in Figure 2 by way of the terminals 41, so as to displace the envelope by an amount corresponding to changes in the sound volume. The reference numeral 77 represents a third DA conversion circuit for converting the digital data from the second adder circuit 76 into an analog voltage. This DA conversion circuit 77 is, like the DA conversion circuit 58, connected between supplies of the source voltage V and the reference voltage such DD that its output varies between these two voltage values.

The reference numeral 78 denotes a noise generating circuit to which the oscillating frequency signal is input and which comprises a shift register, formed by a plurality of flip-flops, and an Exclusive-OR circuit.

The reference numerals 79 and 81 denotes two bell sound ROMs for storing data on the frequency of two such sounds in the form of a rectangular wave. The reference numerals 80 and 82 denote two bell sound programmable counters for dividing the oscillating frequency of the signal from the oscillator circuit 1 in accordance with respective division ratios determined by data stored in the ROM 79 and the ROM 81 and for generating outputs representing the frequencies of the two sounds in rectangular form.The reference numeral 83 denotes a fourth DA conversion circuit, which operates in a manner similar to the second DA conversion circuit 63 and in which the output from the noise generating circuit 78 and the rectangular outputs from the bell sound programmable counters 80 and 82 are mixed and converted into an analog voltage, with the analog voltage output from the third DA conversion circuit 77 as a maximum limiting value. The reference numeral 84 denotes an output terminal for the noise output and the rectangular waves of two different frequencies to which the envelope is added. The reference numeral 86 denotes a switch, which is arranged to be switched on when the data on the sound volume of the rhythmic sound supplied to the input terminal 72 represents a lack of sound.When the switch is switched on, the output from the DA conversion circuit 77 is forced to the reference voltage level. As a result, only the reference voltage is supplied to the DA conversion circuit 83 and no signal representing the rhythmic sound is output to the terminal 84. Further, the output terminals of the noise generating circuit 78 and the two bell sound counters 80 and 81 and the input terminal of the DA conversion circuit 83 are selectively masked in the manufacturing process for the sound synthesizer, which is in the form of an IC. Therefore1 only the desired rhythmic sound is arranged to be output.

Figure 5 illustrates sound volume and pause control circuitry in each of the sound sources and the rhythmic sound generating circuit. Referring to this drawing, the reference numeral 37 denotes the main ROM, and the reference numeral 102 represents an envelope ROM of the sound source or the rhythmic sound generating circuit. The reference numerals 103 to 107 represent an adder for adding data on the sound volume from the main ROM 37 and envelope data from the envelope ROM 102. The portion 107 of the adder in the short dashed line illustrates the addition with respect to one bit.

The portions 103 to 106 are structured similarly. The portion 107 performs addition of the least significant bits and its carry over is input to the portion 106.

Similarly, the carry over of the portion 106 is input to the portion 105, and the carry over of the portion 105 is input to the portion 104. The reference numeral 108 denotes a DA conversion circuit for converting the data representing envelopes into an analog voltage, and the reference numeral 190 denotes a NOR circuit for detecting a state where all the data on sound volume in the main ROM 37 becomes 0 (that is signifying a pause).

The reference numeral 110 denotes a MOS switch for forcing the output of the DA conversion circuit 108 to the reference voltage when the output from the NOR circuit 190 is high. This switch corresponds to the switch 65 shown in Figure 3 and the switch 86 shown in Figure 4.

The operation of the synthesizer is as follows.

The oscillator circuit 1 shown in Figure 1 may comprise a CR oscillator circuit, crystal oscillator circuit or ceramic oscillator circuit so that a signal having a desired oscillating frequency is obtained for input to the sound sources 3, 4 and 5 and the rhythmic sound generating circuit 6. In this instance, the oscillating frequency itself is divided into 1/M for input to the control circuit and the rhythmic sound generating circuit. However, the oscillating frequency may be used as it is without performing the 1/M division, if desired.

The oscillating frequency of the signal supplied to the input terminal 31 of the control circuit shown in Figure 2, from the oscillator circuit 1, is divided into the frequency of the desired tempo by the tempo programmable counter 32. By way of example, it is assumed that the frequency of the input to the tempo programmable counter 32 is 128 Hz and the shortest note desired is , (1/32 times a whole note). In general, if tempo is expressed by J = 60, this signifies a speed (tempo) at which 60 J (quarter note) are generated in a minute so that one J is generated in a second. Since is 1/8 the length of j , 8 may be generated in a second. If the shortest note is a signal frequency at which 8 > can be output needs to be generated by the tempo programmable counter 32.

Thus, the tempo programmable counter 32 must reduce 128 Hz to 8 Hz for the desired result. Therefore, in order to make J = 60, a division ratio of 8/128 = 1/16 is required in the tempo programmable counter 32.

In the case of a tempo programmable counter of five bits, 32 different tempos can be set, since step by step alteration from 1/1 division to 1/32 division can be performed by setting the data stored in the tempo ROM 33 into the values from 00000 to 11111. Therefore, in the above described example, 2 = 60 can be satisfied and tempo setting from J = 30 to ! = 960 can also be realised. By storing a certain number of different tempos, for example 32, in the tempo ROM 33 and making the output from the control ROM 39 select the address of the tempo ROM 33, changes in tempo can be achieved while playing a melody. The output from the tempo programmable counter 32 representing the frequency of the shortest note is supplied to the note programmable counter 34.In the case where the note programmable counter 34, like the tempo programmable counter 32, has five bits and the address of the note ROM 35 is determined in accordance with data on the length of a note output from the main ROM 37, a data output of five bits can be obtained from the note ROM 35. Therefore, 32 different division ratios can be selected by the note ROM 35 for the counter 34. In other words, an output for any note, from the shortest note J to the longest note o (whole note), which is up to 32 times the length of j can be provided.

The clock pulses from the note programmable counter 34 are counted by the main programmable counter 36, and address selection within the main ROM 37 is performed in accordance with the counted value so that the address in the main ROM 37 is incremented. The main ROM 37 stores data on all the notes of a melody (data on note length and pitch, jump data, and data on sound volume and sound separation for each of the sound sources and the rhythmic sound generating circuit).

When the jump data in the main ROM 37 becomes 1, it is counted by the control counter 38 and data stored by the control ROM 39 is output to set or re-set a flip-flop in the main programmable counter 36 whereby data representing the address to which the main ROM 37 is to be jumped is set for the purpose of performing a jump in the read address of the main ROM 37. The control ROM 39 stores the address to which the main ROM 37 is to be jumped so that the next address to which jumping is destined is-selected at every count of the control counter 38.

The output from the tempo programmable counter 32 is supplied to the envelope counter 56 of the sound source shown in Figure 3. The envelope ROM 55 stores data obtained by converting the envelope into a digital value. For example, when a saw shaped envelope is formed by four bits of data, data as shown in Figure 6, is stored in the addresses 0 to 1F of the envelope ROM 55. The main ROM 36 also stores data indicating whether it is necessary to provide a separation at each note and, whenever this separation data is present in the note data, the envelope counter 56 is re-set by the application of short pulses. As a result, the read address of the envelope ROM 55 is also set or re-set to the 0 address. Next, the envelope counter 56 counts the clock pulses generated, at the frequency of'the shortest note, by the tempo programmable counter 32.

The envelope counter 56 successively reads the envelope data from the envelope ROM 55. Data on the sound volume of each of the notes is also stored in the main ROM 37 shown in Figure 2 and is supplied to the input terminal 52 shown in Figure 3 to be added to the read envelope data by the adder circuit 57. Since the envelope data is increased by an amount corresponding to the data representing the sound volume, sound volume adjustment for each of the notes can be performed. The digital data representing the envelope from the adder circuit 57 is converted into an analog voltage by the first DA conversion circuit 58.

The data on the sound pitch from the main ROM 37 shown in Figure 2 determines the address of the pitch ROM 59 shown in Figure 3. Since division ratio data corresponding to the sound pitch is stored in the pitch ROM 59, the division ratio of the pitch programmable counter 60 is thus determined. The pitch programmable counter 60 receives as input the oscillating frequency signal from the oscillator circuit 1 and supplies as output clock pulses having a frequency which is N times the frequency of the sound pitch desired. For example, in a case where a sound pitch of C4 = 256 Hz is desired, assuming that the oscillator frequency is 262.144 kHz, and N = 32, 1/32 division is needed. The output from the pitch programmable counter 60 is supplied to the waveform counter 62 and the address of the waveform ROM 61 is incremented.

The waveform ROM 61 stores data obtained by converting an sound waveform into a digital value. For example, in the case where the number (N) of addresses of the waveform ROM 61 is 32 and the data has a value up to 32, a sine wave as shown in Figure 7 can be stored.

Therefore, when the waveform counter 62 completes its counting, one waveform is output. The frequency of this waveform is the frequency of the sound pitch. In order to change the sound pitch, the frequency of the clock pulses counted by the waveform counter 62 needs to be altered by changing the frequency of the clock pulses supplied by the pitch programmable counter 60. The digital waveform data provided as output from the waveform ROM 61 is supplied to the second DA conversion circuit 63.However, by making the maximum operating voltage of the second DA conversion circuit 63 the analog voltage output from the first DA conversion circuit 58, which has made the above described envelope, and by making the minimum operating voltage a ground voltage, the final output waveform from the second DA conversion circuit 63 becomes the original waveform in analog shape to which an envelope is added.

That is, in the second DA conversion circuit 63, the analog voltage output from the first DA conversion circuit is combined with data output from the waveform ROM 61. Furthermore, when data on sound volume is represented by a 0, the switch 65 is switched on so that output from the sound source is inhibited. This switch may alternatively be connected between the output from the second DA conversion circuit 63 and the source of the reference voltage.

Figure 4 shows the rhythmic sound generating circuit 6. The clock output from the tempo programmable counter 32 shown in Figure 2 is supplied to the rhythm envelope counter 75. The rhythm envelope counter 75 increments the read address of the rhythm envelope ROM 74 in which is stored data obtained by converting the envelope for the rhythmic sound into a digital value.

The digital data from the rhythm envelope ROM 74 is added to data representing the sound volume of the rhythmic sound output from the main ROM 37 shown in Figure 2, and then the result is converted into an analog voltage by the third DA conversion circuit 77.

A signal obtained by 1/M division of the oscillating frequency is supplied from the input terminal 73 to the noise generating circuit 78 and the bell sound programmable counters 80 and 82. As mentioned above, the noise generating circuit 78 includes the multi-stepped shift register, and the Exclusive-OR circuit, which receives outputs from two specific flip-flops of the shift register whereby the output from the Exclusive-OR circuit is arranged to be returned to the first step of the above described shift register. As a result, white noise, in which the maximum frequency is the frequency of the clock pulses input to the shift register, is generated.

The bell sound programmable counters 80 and 82 divide the frequency of the input signal from the oscillator circuit 1 by division ratios output from the bell sound ROMs 79 and 81, which store data on sound frequencies in the form of rectangular waves. As a result, rectangular shaped pulses of desired frequencies are output. The DA conversion circuit 83 mixes the noise from the noise generating circuit 78 and the rectangular pulses having different frequencies from the bell sound programmable counters 80 and. 82. In this state, by making the maximum operating voltage of the DA conversion circuit 83 the analog voltage output from the third DA conversion circuit 77, rhythmic sound pulses to which the envelope is added can be output.

With the noise and the rectangular pulses and the envelopes to be added to the former, rhythmic sounds such as drum, cymbal and bell or the like percussion sounds can be freely synthesized.

Finally, the analog outputs from the plurality of sound sources and the analog output representing the rhythmic sound are mixed by the mixing circuit 7 shown in Figure 1. Figure 8 illustrates the signals generated in the case where a rectangular waveform, sine waveform and saw shaped waveform are stored in the waveform ROMs of the three sound sources, and a saw shaped envelope is stored in each of the envelope ROMs. By storing the waveforms and the envelopes for various instruments in the waveform ROM and the envelope ROM, a variety of sounds can be generated.

As described above, therefore, by providing memories in the sound sources, in which waveforms and envelopes are stored, sound of varying qualities and timbres can be generated by re-writing the data in each memory, whereas in the conventional arrangement only sound of uniform sound quality can be obtained.

Furthermore, by adding sound accentuation data (data on sound volume) to the data on the envelopes, sound volume adjustment corresponding to each note can be achieved.

With this system, without any control by an outside micro-computer, automatic performance of music is possible based on the music information stored in the memories. Furthermore, since a plurality of sound sources, each having individual sound qualities, and in addition the rhythmic sound generating circuit are provided, wide, profound and natural performance canbe realised with respect to the conventional arrangement. Additionally, not only music but natural sounds, such as bird song, insect noise or the like, wind, wave and water, and imitation sounds are also possible.

Claims

1. A sound synthesizer comprising a main store for data providing melody information, and a sound source controlled by data signals output from the main store for generating an output representing melodic sound, the sound source comprising a further store for data regarding the timbre of a sound, a pitch control circuit which, in response to a respective data signal representing sound pitch output from the main store, is arranged to co-operate with the further store so as to generate a sound quality signal representing sound of a selected pitch and timbre, and a sound control circuit responsive to the sound quality signal and to at least one further data signal representing a further sound characteristic output from the main store for providing the output.

2. A synthesizer as claimed in claim 1, wherein the pitch control circuit compr-ises a pitch counter arranged to receive a signal providing a reference frequency and to divide the reference frequency by an amount determined by the data signal representing sound pitch, and wherein the sound source further comprises another counter operable to select a respective address in the further store in dependence upon the count value of the pitch counter for generating the sound quality signal.

3. A synthesizer as claimed in claim 1 or 2, wherein the further store contains data on the waveform of the sound to be produced, and wherein the pitch control circuit is arranged to control the output from the further store of a waveform signal providing the sound quality signal.

4. A synthesizer as claimed in claim 3, wherein the sound control circuit comprises an envelope store for data on an envelope for the waveform, and a circuit for combining an envelope signal generated from the envelope store with the waveform signal.

5. A synthesizer as claimed in claim 4, wherein the combining circuit comprises a first digital to analog converter for converting the envelope signal into an analog voltage signal, and a second digital to analog converter for combining the waveform signal and the analog voltage signal for providing the output.

6. A synthesizer as claimed in claim 4 or 5, wherein the sound control circuit further comprises adding means for adding the at least one further data signal, which further data signal represents sound volume, and a data signal output from the envelope store for generating the envelope signal.

7. A synthesizer as claimed in claim 6, wherein the sound control circuit further comprises a switch coupled between the combining circuit and a source of a reference voltage, and means for operating the switch in dependence upon the further data signal representing sound volume.

8. A synthesizer as claimed in claim 4 or 5, wherein the envelope store is arranged to be re-set to an initial address in response to the at least one further -data signal, which further data signal represents sound separation.

9. A synthesizer as claimed in claim 8, wherein the sound control circuit further comprises an envelope counter for selecting respective addresses in the envelope store in dependence upon the count value thereof, the envelope counter being arranged to count at a rate determined by another data signal representing sound tempo output from the main store.

10. A synthesizer as claimed in claim 1, further comprising a control circuit including the main store, the control circuit further including a tempo counter arranged to receive a signal providing a reference frequency and to divide the reference frequency by an amount determined by a data signal representing sound tempo output from the main store, a sound length counter arranged to receive the output from the tempo counter and to divide the frequency thereof by an amount determined by a data signal representing sound length output from the main store, and a main counter coupled to the sound length counter for selecting respective addresses in the main store.

11. A synthesizer as claimed in any preceding claim, comprising a plurality of the sound sources.

12. A synthesizer as claimed in any preceding claim, further comprising a further sound source for generating an output representing rhythmic sound.

13. A synthesizer as claimed in claim 12, wherein the further sound source comprises means for generating a rhythm signal in the form of pulses having a predetermined frequency.

14. A synthesizer as claimed in claim 12 or 13, wherein the further sound source comprises means for generating a noise signal in the form of random pulses.

15. A synthesizer as claimed in claim 13 or 14, wherein the further sound source comprises a rhythm envelope store for data on an envelope for the pulses, and a circuit for combining a rhythm envelope signal generated from the rhythm envelope store with the pulses.

16. A synthesizer as claimed in claim 15, wherein the combining circuit of the further sound source comprises a first digital to analog converter for converting the rhythm envelope signal into a rhythm analog voltage signal, and a second digital to analog converter for combining the pulses and the rhythm analog voltage signal for providing the rhythmic sound output.

17. A synthesizer as claimed in claim 15 or 16, wherein the further sound source further comprises adding means for adding a data signal representing rhythmic sound volume output from the main store and a data signal output from the rhythm envelope store for generating the rhythm envelope signal.

18. A synthesizer as claimed in claim 17, wherein the further sound source further comprises a switch coupled between the combining circuit of the further sound source and a or the source of a reference voltage, and means for operating the switch in dependence upon the data signal representing rhythmic sound volume.

19. A synthesizer as claimed in any of claims 15 to 18, wherein the rhythm envelope store is arranged to be re-set to an initial address in response to a data signal representing rhythmic sound separation output from the main store.

20. A synthesizer as claimed in claim 19, wherein the further sound source further comprises a rhythm envelope counter for selecting respective addresses in the rhythm envelope store in dependence upon the count value thereof, the rhythm envelope counter being arranged to count at a rate determined by a data signal representing rhythmic sound tempo output from the main store.

21. A synthesizer as claimed in any of claims 11 to 20, further comprising a mixing circuit for mixing the outputs from each of the sound sources.

22. A sound generating device storing melody information in a main storing means thereof, said sound generating device comprising a sound source controlled in accordance with data on the pitch of sound in said melody information read from said main storing means, said sound source comprising sound waveform storing means in which the waveform of sound is stored in a digital manner and the same is repeatedly read therefrom, the frequency of the repetition being the frequency which corresponds to said data on the pitch of sound, envelope waveform storing means in which an envelope waveform is stored in a digital manner and the same is repeatedly read therefrom, and digital to analog conversion means for dividing a voltage formed by digital to analog converting output data from said envelope waveform storing means in accordance with output data from said sound waveform storing means.

23. A sound synthesizer comprising a main store for data providing melody information, and a source of rhythmic sound controlled by data signals output from the main store for generating an output representing rhythmic sound, the rhythmic sound source including means for generating a rhythm signal in the form of pulses, and a rhythmic sound control circuit arranged in response to the rhythm signal and the data signals output from the main store to generate the output representing rhythmic sound.

24. A sound generating device storing melody information in a main storing means thereof, and successively progressing read addresses of said main storing means at a speed corresponding to the length of sound, said sound generating device comprising rhythm sound generating means arranged to be controlled by data on the sound volume stored in said main storing means as said melody information, said rhythm sound generating means comprising noise generating means for outputting pulses at random, pulse outputting means for outputting pulses of a predetermined frequency, envelope waveform storing means for storing, in a digital manner, a rhythm sound envelope waveform, adder means for adding output data from said envelope waveform storing means and said data on the sound volume, digital to analog conversion means for converting an output from said adder means into an analog envelope waveform, and means for adding said analog envelope waveform to a signal output from said noise generating means or said pulse outputting means.

25. A sound synthesizer comprising a main store for data providing melody information, and a sound source controlled by data signals output from the main store for generating an output representing melodic sound, the sound source comprising means responsive to a respective one of the data signals representing sound pitch for generating a digital signal representing sound of a selected pitch, means for generating an envelope signal which is controlled in dependence upon a respective one or respective ones of the data signals representing at least one of sound tempo and sound volume, and a circuit for combining the digital signal and the envelope signal for providing the output.

26. Any novel integer or step, or combination of integers or steps, hereinbefore described and/or as shown in the accompanying drawings, irrespective of whether the present claim is within the scope of or relates to the same, or a different, invention from that of the preceding claims.